Cytoplasmic cyclin D1 regulates cell invasion and

ARTICLE

Received 2 Jul 2015 | Accepted 11 Apr 2016 | Published 16 May 2016

Cytoplasmic cyclin D1 regulates cell invasion andmetastasis through the phosphorylation of paxillinNoel P. Fuste1,*, Rita Fernandez-Hernandez1,*,w, Tania Cemeli1, Cristina Mirantes2, Neus Pedraza1, Marta Rafel1,

Jordi Torres-Rosell1, Neus Colomina1, Francisco Ferrezuelo1, Xavier Dolcet2 & Eloi Garı1

Cyclin D1 (Ccnd1) together with its binding partner Cdk4 act as a transcriptional regulator to

control cell proliferation and migration, and abnormal Ccnd1 �Cdk4 expression promotes

tumour growth and metastasis. While different nuclear Ccnd1 �Cdk4 targets participating in

cell proliferation and tissue development have been identified, little is known about how

Ccnd1 �Cdk4 controls cell adherence and invasion. Here, we show that the focal adhesion

component paxillin is a cytoplasmic substrate of Ccnd1 �Cdk4. This complex phosphorylates a

fraction of paxillin specifically associated to the cell membrane, and promotes Rac1 activation,

thereby triggering membrane ruffling and cell invasion in both normal fibroblasts and tumour

cells. Our results demonstrate that localization of Ccnd1 �Cdk4 to the cytoplasm does not

simply act to restrain cell proliferation, but constitutes a functionally relevant mechanism

operating under normal and pathological conditions to control cell adhesion, migration and

metastasis through activation of a Ccnd1 �Cdk4-paxillin-Rac1 axis.

DOI: 10.1038/ncomms11581 OPEN

1 Cell Cycle Lab, Institut de Recerca Biomedica de Lleida (IRBLleida), and Departament de Ciencies Mediques Basiques; Facultat de Medicina; Universitat deLleida, 25198 Lleida, Catalonia, Spain. 2 Oncopathology Lab, Institut de Recerca Biomedica de Lleida (IRBLleida), and Departament de Ciencies MediquesBasiques; Facultat de Medicina; Universitat de Lleida, 25198 Lleida, Catalonia, Spain. * These authors contributed equally to this work. w Present address: CellCycle Group, Cancer Epigenetics and Biology Program (PEBC), Institut d’Investigacio Biomedica de Bellvitge (IDIBELL), Barcelona, Catalonia, Spain.Correspondence and requests for materials should be addressed to E.G. (email: [email protected]).

NATURE COMMUNICATIONS | 7:11581 | DOI: 10.1038/ncomms11581 | www.nature.com/naturecommunications 1

mailto:[email protected]

http://www.nature.com/naturecommunications

Cyclin D1 (Ccnd1) is a regulatory subunit of the cyclin-dependent kinases Cdk4/6, whose Ccnd1-dependentactivity controls cell proliferation and development

through its role as a transcriptional regulator1,2. Ccnd1 hasbeen associated with tumour invasion and metastasis in clinicalstudies and in in vivo experiments3–5. This association seemsrelated to the ability of Ccnd1 to regulate cell adhesion andmigration, and not to the Ccnd1-dependent mechanisms thatcontrol cell proliferation6. Ccnd1� /� mouse macrophages andfibroblasts show an increment in their capacity to adhere to thecell-matrix, and a reduction in cell motility7,8. These phenotypeshave been attributed to the nuclear role of Ccnd1 as atranscriptional regulator of genes controlling cell adherence andmigration8,9. However, the functional and physical interaction ofCcnd1 with cytoplasmic and membrane-associated proteins, suchas filamin A, PACSIN2, RhoA and Ral GTPases indicate that thiscyclin could play an active role in the cytoplasm regulatingadherence and migration10–14.

The protein paxillin (Pxn) was identified as a tyrosine-kinasesubstrate and described as a structural and regulatory componentof focal adhesions (FAs)15,16, the macromolecular assembliesthrough which the cytoskeleton connects to the extracellularmatrix. The regulation of FA recycling is a key step in the controlof cell adherence and motility17,18. Thus, Pxn-null fibroblastsdisplay abnormal FA formation, delay in cell spreading andreduced migration, while mice deficient in Pxn show earlyembryonic lethality mainly due to the impairment of cellmigration19. In recent years several works have highlighted theimportance of Pxn not only as an organizer of FAs but also as amolecular scaffold coordinating different signalling pathways20.Pxn serves as a substrate for several serine-threonine kinases inresponse to adhesion stimuli and growth factors21–24, beingregulated and playing functional roles at cellular locations distinctfrom FAs such as membrane ruffles25,26.

Ccnd1� /� cells spread more rapidly, show an elevatednumber of adhesions sites centripetally distributed around thecircumference of cells, and also exhibit augmented levels oftyrosine-phosphorylated Pxn7,8, suggesting the existenceof alterations in the adhesion machinery. In this work, weshow that the Ccnd1 �Cdk4 complex phosphorylates asubpopulation of Pxn present in membrane ruffles but not inFAs, which is functionally relevant in the control of cell spreadingand invasion in both normal fibroblasts and tumour cells.Although it is widely accepted that the accumulation of Ccnd1 inthe cytoplasm operates only as a sequestration mechanism toprevent cell proliferation27,28, our results demonstrate thatcytoplasmic Ccnd1 has an active role in the induction of cellmigration and invasion. In addition, the existence of aCcnd1 �Cdk4-Pxn-Rac1 axis helps explain the invasiveproperties of tumours overexpressing Ccnd1.

ResultsPxn binds to and is an in vitro substrate of Ccnd1 .Cdk4.Depletion of Ccnd1 promotes cell attachment to the extracellularmatrix, a process likely mediated through the stabilization ofFAs8. Considering that FAs are central elements to the control ofcell adherence and migration, we explored whether Ccnd1 couldinteract with FA components. In mouse fibroblasts, we foundspecific co-immunoprecipitation (co-IP) of both endogenousCcnd1 and Cdk4 with Pxn (Fig. 1a), a key component of FAs20.In Ccnd1� /� fibroblasts we were unable to co-IP Cdk4, eventhough the amount of immunoprecipitated Pxn was slightlyhigher than in wild-type cells. This likely indicates that Cdk4must form a complex with Ccnd1 in order to interact with Pxn.IP of endogenous Ccnd1 also brought down Pxn in a specific way,

albeit the amount of immunoprecipitated Pxn was very lowcompared with total Pxn in the whole cell extract (Fig. 1b). Ourresults are compatible with a significant amount of Ccnd1interacting with a small fraction of total Pxn in the cell (seebelow). In addition, we have also observed the interactionbetween Pxn and Ccnd1 under heterologous conditions. We co-transfected both green fluorescent protein (GFP)-Pxn and Flag-Ccnd1 into rat prostate tumour cells (R3327-50A) and performedan IP against GFP. The anti-GFP antibody was able to co-IP theflag-tagged Ccnd1 only when GFP was fused to Pxn, but not incontrol cells transfected with GFP alone (Fig. 1c). In order to testwhether the interaction between Ccnd1 and Pxn is direct, wecarried out in vitro GST-pull down assays. GST-fusions with full-length Pxn or only with the C-terminal domain of the proteinpurified from bacteria were mixed with Ccnd1 produced byin vitro translation. We recovered Ccnd1 bound to glutathionebeads only when the fusion constructs were used, but not withGST alone (Fig. 1d). Overall, our results indicate that there is aspecific and direct interaction between Pxn and Ccnd1 �Cdk4 atendogenous levels in unperturbed cells.

Pxn is regulated by phosphorylation at different residues inresponse to a plethora of extracellular stimuli20. Because Pxncontains many putative Cdk-phosphorylation sites, we analyzedwhether Pxn serves as a substrate for the Ccnd1 �Cdk4 complex.Ccnd1 �Cdk4 complexes purified from insect cellsphosphorylated GST-Pxn obtained by heterologous expressionin E. coli (Fig. 1e). Omission of the Ccnd1 �Cdk4 complex orusing the Cdk4/6 specific inhibitor Palbociclib preventedphosphorylation of GST-Pxn, confirming that the observedphosphorylation was due to the Ccnd1 �Cdk4 complex includedin the assay. To pinpoint the phosphorylated residues, we firststudied the in vitro phosphorylation of deleted constructs, andnext we created point mutations by site-directed mutagenesis.The analysis of these mutant versions of Pxn by in vitrophosphorylation allowed us to establish that Ccnd1 �Cdk4 targetsthree different serines (S83, S178 and S244) in Pxn (Fig. 1f). Inaddition, we confirmed the phosphorylation at serine 83 by massspectrometry (Supplementary Fig. 1A; Supplementary Tables 1and 2). Failure to phosphorylate the mutated versions was notdue to the lack of interaction, because we were still able to co-IPcomparable amounts of hemagglutinin (HA)-tagged Ccnd1 withwild-type and mutant versions of GFP-tagged Pxn in co-transfected human HEK293T cells (see Supplementary Fig. 1B).Whereas the S244 residue is within a consensus sequence for theCdk2 kinase, and it is phosphorylated by Cdk5 duringoligodendrocyte differentiation24, phosphorylation of Pxn atserines 83 and 178 has been involved in the regulation of celladhesion and migration. As Ccnd1 has a role in the control of celladhesion and migration7,8, we have centred our study in theimportance of phosphorylation at serines 83 and 178.

Pxn phosphorylation by Ccnd1 .Cdk4 in invasion and spreading.Ccnd1-deficient fibroblasts show the same diameter size thanwild-type cells, but attach and spread more rapidly than theseafter seeded in fibronectin-coated plates7,8. Since Pxn is requiredfor efficient and rapid spreading of fibroblasts in fibronectin19, wehypothesized that Ccnd1 could negatively regulate cell spreadingthrough the phosphorylation of serines 83 and 178 in Pxn. Inorder to test this, we carried out functional assays with single anddouble phosphomimetic (serine to glutamic acid) and non-phosphorylatable (serine to alanine) Pxn mutants (see Figs 2and 3). First, we transfected these mutants fused to GFP intoCcnd1� /� fibroblasts, and green cells were evaluated for theirspreading capacity (Fig. 2a,b). Under our assay conditions,expression of Ccnd1 produced a delay of spreading in otherwise

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11581

2 NATURE COMMUNICATIONS | 7:11581 | DOI: 10.1038/ncomms11581 | www.nature.com/naturecommunications


Ccnd1� /� fibroblasts. This effect was mimicked by the singlephosphomimetic S83E allele (as well as the double mutantS83,178E) (Fig. 2b and Supplementary Fig. 2A). However, boththe S83E S178A and the S83A S178E alleles with a non-phosphorylatable residue at position 178 or 83, respectively, didnot delay spreading (see also Supplementary Fig. 2B). Hence,phosphorylation at both sites is required for Pxn-dependentcontrol of cell spreading. Presumably, a kinase other thanCcnd1 �Cdk4 must be responsible for the phosphorylation ofserine 178 when Ccnd1� /� cells are transfected with the singlemutant S83E. By contrast, the single S178E mutant only had aneffect on spreading when co-transfected with Ccnd1. Thisstrongly suggests that the phosphorylation event of serine 83that is relevant for the spreading effect depends on Ccnd1.

We also analyzed the importance of Ccnd1 �Cdk4 in regulatingthe spreading of rat prostate tumour cells (R3327-50A), whichshow an enhanced metastatic potential29. Downregulation ofCcnd1 by RNA interference significantly augmented the ability ofR3327-50A cells to spread in fibronectin-coated plates(Supplementary Fig. 3A,B). Expression of mutant alleles of Pxnin Ccnd1-deficient R3327-50A cells produced similar results tothose just described for fibroblasts (Fig. 2c), indicating thatphosphorylation of Pxn by Ccnd1 �Cdk4 is also important in theregulation of spreading in these cells. Expression of theCcnd1K112E mutant, which cannot produce activeCcnd1 �Cdk4 complexes, did not rescue the deficiency of Ccnd1in R3327-50A cells, corroborating that Ccnd1 �Cdk4 kinaseactivity is required for cell-spreading control.

Ccnd1

Pxn

Cdk4

Input

Ccnd1+/++/+ Ccnd1–/–

IgG IgG

IP

IP

Inputanti

-Ccnd1IgG

Ccnd1

Pxn

Ccnd1

Input IP: anti GFP

WB

: ant

i GF

P

GFPPxn

GFP

flag-Ccnd1

Ab HC

GFPPxn

GFP GFPPxn

GFP

Input Pull downan

ti-G

ST

anti-

Ccnd1

GST-Pxn-Ct

Ccnd1

– – –

+ +–

+

–

+ – + +GSTPxn

GST

D1/Cdk4

+ + – –– +– ++ ++ –

–++

Palbociclib – + – –+

GST

– – –

+ +–

+

–

+ – + +

– –+ –– –+ –

GST

* *

32P

Input

GST-Pxn

S83AS178AS244Awt S83A

S83AS178A

S178AS244A

S83AS244AS244A stS178A

32P

Input

0.64±0.02

0.54±0.03

1,00±0.05

0.48±0.08

0.33± 0.17

0.08±0.06

Ratio 0.43±0.11

97Kd

66kd

45kd

45kdanti flag

30kd

30kd

30kd

66kd

30kd

66kd

97kd

30kd

97kd

66kd

45kd

0.47±0.07

GST-PxnGST-pRB

R3327-5′A cells+ flag-Ccnd1

Pxn-Ct

GST-Pxn

–/–

anti-Pxn

anti-Pxn

a b

c d

fe

Figure 1 | Pxn directly binds to and is an in vitro substrate of Ccnd1 .Cdk4. (a) A rabbit polyclonal antibody (anti-Pxn) was used to IP endogenous Pxn from

Ccnd1� /� and Ccnd1þ /þ fibroblasts. A rabbit polyclonal antibody against the Flag epitope (IgG) was used as a mock experiment. Input and IP samples were

analyzed by western blot to detect Ccnd1, Cdk4 and Pxn. (b) IP with a rabbit polyclonal anti-Ccnd1, and anti-Flag (IgG) as a mock control in wild-type fibroblasts.

(c) Rat prostate tumor cells R3327-50A were co-transfected with GFP-tagged human Pxn, or an empty GFP vector, and Flag-tagged human Ccnd1. Cell lysates

were immunoprecipitated with an anti-GFP monoclonal antibody, and immunoblotted with anti-GFP (top panel) or anti-Flag (bottom panel). Ab HC, antibody

heavy chain. (d) The Ccnd1 protein produced by in vitro translation was incubated with GSTor GST-Pxn fusion proteins (full length or C-terminal region containing

the four LIM domains, aa337–591) purified from E. coli. Input and pull down samples were analyzed by western blot to detect Ccnd1 and GST. Asterisk indicates

degradation bands. (e) Ccnd1 �Cdk4 complexes (Sigma) were assayed for kinase activity against GST-Pxn (full length) and GST-Rb1 (aa379–928). The Cdk4/6

inhibitor Palbociclib was added at 2mM. Coomassie blue staining was used to test equal loading (bottom panel). (f) Kinase assay as in E for different non-

phosphorylatable mutants of Pxn. Band intensity was quantified with ImageJ under conditions of unsaturated signal exposure. The phosphorylation efficiency is

shown as a ratio relative to wild type (mean±s.d.) of two independent experiments; st, size standard.

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11581 ARTICLE



Ccnd1-deficient cells migrate and invade less than wild-typecells8,10. Since phosphorylation of Pxn at serines 83 and 178 isrequired for efficient migration22,23, we postulated thatCcnd1 �Cdk4 may exert its positive effect on migration andinvasion through Pxn phosphorylation. We analyzed thecontribution of Ccnd1 to the ability of R3327-50A cells toinvade in matrigel-coated transwells. Downregulation of Ccnd1by short hairpin RNA (shRNA) dramatically reduced the invasioncapacity of these cells, but this was restored by the expression ofsingle and double phosphomimetic mutants (Fig. 3). Contrary towhat we observed in the spreading assays, both singlephosphomimetic alleles (HA-Pxn S83E and HA-Pxn S178E)

rescue the invasion capacity of Ccnd1-depleted cells (Fig. 3a),which suggests that Ccnd1 may regulate invasion through thephosphorylation of Pxn at serine 83 and 178.

Reduced Pxn S83 phosphorylation in Ccnd1-deficient cells.Our functional assays indicate a prominent role of Ccnd1, via Pxnphosphorylation, in the regulation of both cell spreading andinvasion. Particularly, our in vitro phosphorylation assays andfunctional results suggest that in vivo Ccnd1 �Cdk4 phosphor-ylates Pxn at serine 83. Hence, we used a phospho-specific anti-body to compare the levels of phosphorylated serine 83 in Pxnbetween wild-type and Ccnd1-deficient cells (Fig. 4a). Surpris-ingly, although we consistently observed a modest decrease(30–40%) in the phosphorylation of serine 83 in the absence ofCcnd1, there was still an important contribution to serine 83phosphorylation that was independent of Ccnd1. This was truefor Ccnd1� /� fibroblasts, R3327-50A tumour cells whereinCcnd1 was knocked down with RNA interference (see alsoSupplementary Fig. 3C), or Ccnd1þ /þ fibroblasts treated with2 mM Palbociclib, a specific inhibitor of Ccnd1 �Cdk4 (Fig. 4a).

ERK1/2 kinase phosphorylates Pxn at serine 83 in FAs23. Tobetter highlight the contribution of Ccnd1 to serine 83phosphorylation, we examined Pxn phosphorylation in thepresence of the specific inhibitor of the extracellular-signal-regulated kinases (ERK) pathway activity U0126. Wild-type andCcnd1� /� fibroblasts were deprived from serum to bring to aminimum both Ccnd1 levels and ERK activity. After 24 h, cellswere refed with serum containing 10mM U0126, and at differenttime points samples were recovered for analysis by western blot(Fig. 4b). Quantification of phosphorylated Pxn at serine 83 versustotal Pxn showed a much higher ratio of Pxn phosphorylation inCcnd1þ /þ cells than in Ccnd1� /� cells, particularly after 6 h ofincubation in serum when Ccnd1 had been clearly induced(Fig. 4c,d). Moreover, we assessed the role of Ccnd1 in promotingPxn phosphorylation under conditions similar to the spreadingassay. R3327-50A rat cells previously infected with shRNA againstCcnd1 (shD1) were transfected with vector or with human Ccnd1and were seeded in fibronectin-coated plates for 2 h. These cellswere also incubated in the absence of serum and treated withDMSO or 10mM U0126. Under these conditions, the expression ofCcnd1 promoted phosphorylation of Pxn at S83 even in theabsence of ERK activity (Fig. 4e,f). Overall, these results areconsistent with Ccnd1 �Cdk4 phosphorylating a subpopulation oftotal Pxn, which may be functionally important in the regulation ofcell spreading and invasion. Also, these results show thatCcnd1 �Cdk4 and ERK can act independently on Pxn regulation.

Cyclin D1 co-localizes with Pxn in the cell membrane. Becauseour previous experiments had suggested that Ccnd1 may bind toand phosphorylate just a specific pool of cellular Pxn, we rea-soned that Ccnd1–Pxn interaction may be constrained by theirsubcellular localization. This prompted us to study the localiza-tion of Ccnd1, Pxn and phospho-serine 83 Pxn by immuno-fluorescence and confocal microscopy. To this purpose,fibroblasts seeded on fibronectin-coated plates were incubated for3 h and then fixed. As expected, we detected accumulation ofCcnd1 in the nucleus of most cells, but many also showed adiffuse cytoplasmic signal (Fig. 5a). Importantly, this diffusesignal was specific as judged by the almost complete absence offluorescence in Ccnd1� /� cells (Supplementary Fig. 4A).Interestingly, about 30% of cells showed co-localization of Ccnd1with Pxn in the cell membrane. By contrast, we did not observeco-localization of Ccnd1 and Pxn at FAs. We also observed co-localization of Ccnd1 with S83-phosphorylated Pxn along the cellmembrane of fibroblasts (Fig. 5b). The same result was obtained

GFP

Ccnd1

Pxnwt

Pxnwt

Ccnd1

Pxnwt

Ccnd1k112

PxnS83AS178A

PxnS83AS178ACcnd1

PxnS83E

S178E

0.00

0.10

0.20

0.30

0.40

0.50

0.60

Rat

io o

f spr

ead

cells

b**

ns

c

GFP

** *

****

*

ns

ns

ns

ns ns

**

**

Pxnwt

Pxnwt

Ccnd1

PxnS83ES178E

PxnS83E

PxnS83ECcnd1

PxnS178E

PxnS178ECcnd1

PxnS83ES178A

PxnS83AS178E

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

Rat

io o

f spr

ead

cells

Ccnd1–/– fibroblasts

R3327-5′A cells + shRNA D1

GFP-PxnS83E S178E

GFP-PxnS83A S178AGFP-Pxn wt

Ccnd1–/– fibroblasts

a

Figure 2 | Pxn phosphorylation at serines 83 and 178 is required for

Ccnd1-dependent delay in cell spreading. (a) Representative image of

unspread and spread morphology in fibroblasts after transfection with

various alleles of GFP-Pxn. (20 mm bar) (b,c) Ccnd1� /� fibroblasts (b) or

R3327-50A cells expressing the shRNA �D1 (c) were co-transfected with an

allele (wild type or mutant) of GFP-Pxn and either HA-Ccnd1 or an empty

vector. Forty-eight hours after transfection, cells were trypsinized and

seeded in serum-free medium in 35-mm well plates coated with 5 mg ml� 1

fibronectin. Thirty minutes (fibroblasts) or one hour later (R3327-50A cells),

the proportion of spread green cells was determined (see Methods for

details). Data (mean±s.e.m.) are from three or more independent

experiments. Significance values were determined by one way ANOVA and

Tukey-HSD post-test (*Po0.05, **Po0.01; ns, no significant).




in tumour cells using an HA-tagged version of Ccnd1(Supplementary Fig. 5).

Neither Ccnd1 nor Pxn seem to be homogeneously distributedthroughout the cell membrane, but rather to specific regions of it.This is consistent with previous findings that associate bothCcnd1 and Pxn to membrane ruffles10,13,26. Rac1 is essential forthe formation of membrane ruffles and can be used as a markerfor these structures30. Hence, to determine if Ccnd1 localized tomembrane ruffles, we studied the co-localization of Ccnd1 withRac1. Again, about one-third of cells showed co-localization ofCcnd1 with Rac1 in the cell membrane, indicating that the

interaction between Ccnd1 and Pxn very likely takes place atthese ripples of the membrane (Supplementary Fig. 4B).Furthermore, membrane ruffles are regions with activemembrane recycling and are thus enriched in the transferrinreceptor protein (TFR)31. Accordingly, we examined co-localization of Ccnd1 with TFR in our cells. Once more, co-localization was only observed at specific regions in the cellmembrane (Supplementary Fig. 4C), supporting our conclusionof a specific localization of Ccnd1 to membrane ruffles.

Finally, we checked whether the localization of Pxn was alteredin Ccnd1� /� fibroblasts. By immunofluorescence, both total and

Control

shD1

Pxn wt PxnS83ES178E

PxnS83E

S178EshD1

PxnS83E

PxnS83EshD1

PxnS178E

0

100

200

300

400

500

600

700

Inva

ding

cel

ls

b

Con

trol

shD

1

Input Matrigel Input Matrigel

shD1

PxnS83 E

HA-Pxn

Ccnd1

shD1

Control

shD1

PxnS83,178 E

HA-Pxn

Ccnd1

Tubulin

Pxnwt

shD1 shD1

ControlPxn

S178 E

shD1

66kd

30kd

66kd

30kd

45kd

a**

**ns ns

*

* ** *

Pxn

S83

E

Pxn

S83

E+

shD

1

Pxn wtshD1

PxnS178EshD1

Gel stain

c

R3327-5′A cells

Figure 3 | Ccnd1 .Cdk4 regulates invasion through the phosphorylation of Pxn. (a) Prostate tumour cells (R3327-50A) were infected with interference

shRNA against Ccnd1 (shD1, Sigma) or with a scramble shRNA as a control. These cells were further infected with a wild type HA-Pxn, an S83, 178E HA-

Pxn, an S83E HA-Pxn, an S178E HA-Pxn or with an empty vector, and 5� 104 co-infected cells were seeded in 24-well transwell filters previously coated

with matrigel, and allowed to invade for 24 h (see Methods for more details). Relative values are expressed as mean±s.e.m. Data are from three

independent experiments. Significance values were determined by one way ANOVA and Tukey-HSD post-test (*Po0.05, **Po0.01, ns no significant).

(b) Representative images (50 mm bar) of the experiment in A. Cells were fixed and stained with Hoescht (input). Non-invading cells were removed using

a cotton applicator (matrigel). (c) Immunoblots showing the expression of Pxn (rat anti-HA) and Ccnd1 (monoclonal antibody DCS6) in the co-infected

cells. Tubulin or gel staining were used to test equal loading.




serine-83-phosphorylated Pxn were greatly reduced in the cellmembrane in both immortalized and primary Ccnd1� /� mouseembryonic fibroblasts (MEFs) (Fig. 5c,d; Supplementary Fig. 6A,B).By contrast, FAs retained a considerable amount of phosphory-lated Pxn in these cells. This is consistent with the modest decreasein phosphorylation detected in western blots (see above, Fig. 4a).Therefore, it seems likely that Ccnd1 �Cdk4 regulates only thefraction of Pxn located in membrane ruffles. Hence, we analyzedthe localization of Ccnd1 and Pxn by cell fractionation. We

obtained a soluble fraction containing cytosolic and nuclear solubleproteins, and a membrane-enriched fraction as indicated byprotein markers for each fraction (Supplementary Fig. 7A). BothCcnd1 and Pxn were found in the membrane fraction of fibroblasts(Supplementary Fig. 7A). To determine whether Ccnd1 and Cdk4form active complexes in the membrane, we immunoprecipitatedCcnd1 from membrane fractions of R3327-50A cells (which havehigher endogenous levels of Ccnd1), and performed a kinase assaywith those fractions using GST-Pxn as substrate. As shown in

VectorU0126

Ccnd1U0126 Vector

Ccnd1

0.00

0.20

0.40

0.60

0.80

1.00

Rel

ativ

e ra

tio o

f pho

S83

vs

tota

l Pxn

at 2

h a

fter

seed

ing

in F

N w

/o s

erum

Ccnd1+/+

Ccnd1–/–

0.00

0.20

0.40

0.60

0.80

1.00

1.20

Rel

ativ

e ra

tio o

f pho

S83

vs

tota

l Pxn

at 6

h a

fter

refe

edin

g Ccnd1+/+

Ccnd1–/–

0

0.1

0.2

0.3

0.4

0.5

0 1.5 3 4.5 6

Rat

io o

f ph

oS83

vs to

tal P

xn

Time (h)

Serum + U0126 10μM

a

c

+/+ –/–

PxnphoS83

Pxn

1 0.69±0.03

Ccnd1

DMSO Palb

1 0.61±0.15

Ccnd1+/+

PxnphoS83

Pxn

ERK

Ccnd1

ERKP42/44

1.5 3 60 1.5 3 60 h

Ccnd1+/+ Ccnd1–/–

d

Refeeding Refeeding

Serumo/n

b

**

Ccnd1

*

* *

Fibroblasts

f

**

**

PxnphoS83

Pxn

ERK

ERKP42/44

GFP-Ccnd1

Vector Ccnd1 Vector Ccnd1

e

**

66kd

30kd

66kd

30kd

45kd

45kd66kd

66kd

66kd

45kd

45kd

66kd

66kd

Serum + U0126 10 μM

U0126 10 μM

FN 120 min w/o serum

R3327-5′A cells+shD1 R3227-5′A cells + shD1

Ccnd1+/+U0126

Ccnd1–/–U0126

Relative ratiophoS83/ total Pxn

Figure 4 | Ccnd1 knock-down leads to a reduction of phosphorylated Pxn at serine 83. (a) By immunoblot densitometry with the Image-Lab 4.0.1

software from BioRad, Pxn phosphorylated at serine 83 (phoS-83 Pxn) and total Pxn levels were determined in Ccnd1� /� and Ccnd1þ /þ fibroblasts, and in

Ccnd1þ /þ cells treated with 2 mM Palbociclib. Equal amounts of total protein per lane were loaded. Data are expressed as mean±s.e.m. (n¼ 3).

Significance was determined by a t-test. (b) Accumulation of phoS-83 and total Pxn in Ccnd1� /� and Ccnd1þ /þ fibroblasts after serum refeeding. At time

zero cells were treated with 10 mM U0126, a specific inhibitor of the ERK pathway. Total ERK, P42/44 ERK, and Ccnd1 were examined in the same

membranes. Wild-type fibroblasts cultured with serum were used as control. Asterisk indicates a lower exposure of the same membrane. (c) Quantification

of the proportion of phosphorylated Pxn versus total Pxn from (b), by immunoblot densitometry as in (a). (d) The proportion of phosphorylated Pxn versus

total Pxn at six hours after refeeding is plotted. Data are expressed as mean±s.e.m. (n¼4). Significance was determined by one way ANOVA and Tukey-

HSD post-test (*Po0.05, **Po0.01, ns not significant). (e) Accumulation of phoS-83 and total Pxn in R3327-50A cells seeded in fibronectin. Cells,

previously infected with shRNA against Ccnd1 (shD1), were transfected with vector or with Ccnd1, and were seeded in fibronectin-coated plates for 2 h in

the absence of serum, and treated with DMSO or U0126. Total ERK, P42/44 ERK, and Ccnd1 were examined in the same membranes. (f) The proportion of

phosphorylated Pxn versus total Pxn is plotted. Samples without ERK inhibitor and with inhibitor were loaded in different gels, but quantification was made

relative to the same sample (expressing Ccnd1 and without U0126) loaded in all the gels. Data are expressed as mean±s.e.m. (n¼4). Significance was

determined by one way ANOVA and Tukey-HSD post-test (*Po0.05, **Po0.01; ns, not significant). Quantification by densitometry as in a.




Supplementary Fig. 7, purified Ccnd1 �Cdk4 complexes from thecell membrane are active, at least as measured by our in vitroassays.

Rac1 activity is downregulated in Ccnd1-deficient cells. Tovarying degrees, Ccnd1� /� cells show more spread morphologythan the corresponding wild type and also exhibit augmentednumber of FAs7,8, with higher levels of tyrosine-phosphorylatedPxn. In contrast, we have observed less accumulation of Pxn inthe membranes of Ccnd1� /� cells in this work (Fig. 5c). Thisdecrease could be in fact a consequence of morphological changesin these cells, such as a reduction of membrane ruffles. Rac1GTPase is the major inductor of membrane ruffling and isrequired for cell migration and invasion26. Pxn induces migration

and Rac1 activation through several mechanisms that promotethe recruitment of Rac1-associated guanine nucleotide exchangefactor (GEF) activity to the leading edge of the cells20. SinceCcnd1 �Cdk4 regulates Pxn phosphorylation and cell invasion, wehypothesized that Ccnd1 �Cdk4 could alter Rac1 activity via Pxnphosphorylation. Then, a reduction on Ccnd1 levels could lead toa reduction of membrane ruffles, and a concomitant decline incell invasion. Therefore, we examined both the localization andactivity of Rac1 in Ccnd1� /� cells. We transfected wild type andCcnd1� /� fibroblasts with a yellow fluorescent protein-Pak1binding domain (YFP-PBD) construct that acts as a fluorescentbiosensor of Rac1 activity32. Forty-eight hours after transfection,we seeded the cells in fibronectin for 1 h, then fixed them,and processed for immunofluorescence (IF). Although sometransfected cells showed strong diffuse YFP or nuclear signal, a

Ccnd1+/+

Ccnd1–/–

0

0.1

0.2

0.3

0.4

0.5

Rat

io o

f cel

ls w

ith p

axill

inac

cum

ulat

ed in

ruf

fles

Ccnd1Pxn Merge Fibroblasts

Z a

xis

Ccnd1 phoS83 Pxn Merge Fibroblasts

Ccn

d1+

/+C

cnd1

–/–

Pxn Merge Fibroblasts

Z a

xis

Z a

xis

phoS83 Pxn

Ccnd1phoS83 Pxn

Ccnd1–/–

a

b

c d

Figure 5 | Ccnd1 co-localizes with Pxn in membrane ruffles of fibroblasts and tumour cells. (a) Fibroblasts were fixed in 4% paraformaldehyde and

permeabilized with 0.2% Triton X-100. Images were acquired by confocal microscopy (10 mm bar). Nuclei were stained with Hoescht (blue). Anti-Ccnd1

(rabbit monoclonal clone EP12) and anti-Pxn (mouse monoclonal clone 349) antibodies were used. (b) Images of cultured Ccnd1� /� and Ccnd1þ /þ

fibroblasts were processed as in A except for permeabilization (low conditions, 0.02% Triton X-100) (10mm bar). Anti-Ccnd1 (mouse monoclonal clone 72-

13 G) and anti-Pxn (S83) phospho-specific (rabbit polyclonal) antibodies were used. Note that the phospho-specific antibody against Pxn gives a nuclear

signal that must be non-specific because total Pxn shows exclusion from the nucleus. (c) Images of cultured Ccnd1� /� and Ccnd1þ /þ fibroblasts were

processed as in a (10mm bar). Primary antibodies were the same as in a (anti-Pxn) and b (anti-Pxn S83). (d) Ratio of cells displaying Pxn accumulation in

membrane ruffles, analyzed from the images in c (see also Supplementary Fig. S6 and Methods). The number of counted cells was nZ179. Bars indicate the

confidence limits for a proportion (a¼0.05).




clear accumulation of YFP-PBD signal in the membranes of wild-type fibroblasts (Fig. 6a) was generally observed. This signal co-localizes with Rac1 and Ccnd1 (Supplementary Fig. 8).

Importantly, few Ccnd1� /� cells showed YFP signal inmembranes (Fig. 6a,b). A similar result was obtained byanalyzing total Rac1 in immortalized and primary MEFs

Vector

Ccnd1Rac1Q61L

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

Ratio

of sp

read c

ells

Control

shD1

Rac1Q61L

Rac1 Q61LshD1

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

Rel

ativ

e ra

tio o

f inv

adin

g ce

lls

Ccnd1+/+

Ccnd1–/–

0

0.1

0.2

0.3

0.4

f

GTP

Total

WB:anti-rac1

+/+ –/–

Fibroblasts

Ccnd1

Beads

Ratio GTPvs total

1 0.60±0.11

*

*

ns

Pxnwt

Pxnwt

PxnS83,178A

WB:anti-rac1

+– Ccnd1+

g

Ratio GTPvs total

1±0.10

0.63±0.12

*

0.59±0.03

Beads

Ccnd1+/+ Ccnd1–/–

Ccnd1+/+6050403020

Gre

y va

lue

100

Fib

robl

asts

GTP WB:anti-rac1

Beads

Pxnwt

PxnS83,178E

Ratio GTPvs total

1±0.06

0.64±0.02

*

GFP-Pxn

30kd

30kd

30kd

st

Total

R3327-5′A+shD1

GT

PT

otal

R3327-5′A +shD1

Ra

tio o

f ce

lls w

ith Y

FP

sig

na

l in

th

e m

em

bra

ne

Ccnd1 –/– fibroblasts

R3327-5′A cells

Distance0 10 20

Ccnd1–/–6050403020

Gre

y va

lue

100

Distance0 10 20

YFP-PBD

ba

c d e

Figure 6 | Ccnd1 knock-down cells show reduced levels of Rac1 activation in the membrane. (a) Ccnd1� /� and Ccnd1þ /þ fibroblasts transfected with

YFP-PBD were seeded in fibronectin-plates for 1 h and then fixed 10 min on ice in 2% PFA to avoid YFP-signal loss (10 mm bar). We measured YFP signal

with Image J. Two representative measures are shown. (b) Ratio of cells displaying YFP-signal accumulation in membrane calculated from two independent

experiments (mean±s.d.). Cells counted: nZ 200. (c) Quantification of Rac1 activity in Ccnd1� /� and Ccnd1þ /þ fibroblasts by Rac1-GTP pull-down

assay. Values show densitometric analysis of relative activity of Rac1 normalized for whole cell lysates (mean±s.e.m.; n¼4). Significance was determined

by a t-test (*Po0.05). (d) R3327-50A cells previously infected with shRNA against Ccnd1 (shD1, Sigma) and transfected with Pxn or with Pxn S83,178A,

and human Ccnd1 were used in Rac1 pull-down assays. Relative values are expressed as mean±s.e.m. (n¼4). Significance was determined by a t-test

(*Po0.05). (e) The same cells as in d were transfected with wild type Pxn or with the phosphomimetic Pxn S83,178E mutant, and were used in Rac1 pull-

down assays. Relative values are expressed as mean±s.e.m. (n¼ 3). Significance was determined by a t-test (*Po0.05). (f) R3327-50A cells were infected

first with shD1 or with a scramble shRNA as a control, then they were further infected with the hyperactive allele Rac1Q61L or with an empty vector, and

cells were seeded in 24-well transwell filters previously coated with matrigel, and allowed to invade for 24 h. Relative values are expressed as mean±s.e.m.

(n¼ 3). Significance was determined by one way ANOVA and Tukey-HSD post-test (*Po0.05; ns, no significant). (g) Ccnd1� /� fibroblasts were

transfected with Rac1Q61L or HA-Ccnd1 or an empty vector. Forty-eight hours after transfection, cells were seeded in serum-free medium in fibronectin-

coated plates. Thirty minutes later the proportion of spread green cells was determined and plotted. Data are from two independent experiments and

expressed as mean±s.d.




(Supplementary Fig. 9). In addition, we detected a drop ofactivated Rac1 (by 40%) in Ccnd1� /� fibroblasts, even thoughthe total levels of Rac1 remained unchanged in these cells(Fig. 6c). These results strongly suggest that Rac1 is active inmembranes of normal fibroblasts during spreading but not inCcnd1� /� cells. In R3327-50A rat tumour cells that had Ccnd1downregulated by RNA interference, activated Rac1 levelsdepended on Ccnd1 because transfecting these cells withhuman Ccnd1 restored the wild-type levels of activated Rac1.However, co-transfecting human Ccnd1 with a non-phosphorylatable allele of Pxn (S83,178A) had no effect on thelevels of activated Rac1 (Fig. 6d and Supplementary Fig. 9E). Inaddition, the expression of a phosphomimetic allele of Pxn(S83,178E) restored Rac1 activation in Ccnd1 knock-down cells(Fig. 6e). Hence, these results indicate that activation of Rac1GTPase by Ccnd1 is mediated by Pxn phosphorylation. Also, thedecrease in Rac1 activation might be responsible for themorphological alterations in Ccnd1� /� cells.

If Ccnd1 exerts its effects on migration through a pathway thatleads to Rac1 regulation, then hyperactivation of Rac1 shouldrescue the invasion phenotype of Ccnd1-deficient cells. Conse-quently, we tested whether expression of a hyperactivated allele ofRac1 (Rac1Q61L) was able to recover the invasion potential ofCcnd1-compromised R3327-50A cells. We found that indeed thisis the case (Fig. 6f). Also, the expression of Rac1Q61L produced adelay in the spreading of Ccnd1� /� fibroblasts comparable tothose transfected with Ccnd1 (Fig. 6g). Taken together theseresults suggest that Ccnd1 regulates cell migration in a cascade ofevents that lead to Rac1 activation through Pxn phosphorylation.

Phospho-Pxn restores metastases by Ccnd1-deficient cells.Ccnd1 is a marker of poor prognosis and has been associated withmetastasis in clinical studies3. Ccnd1-deficient cells consistentlyshow a reduced metastatic potential4,5. Because our results pointto Pxn as a mediator of the effects due to Ccnd1 on cell adherenceand migration, we wanted to test whether a phosphomimeticversion of Pxn was able to rescue the low metastatic potential ofCcnd1-deficient cells. To this end, we performed a metastasisassay in vivo by bloodstream injection of R3327-50A cells in 12-weeks-old nude mice. Animals were euthanized 2 weeks afterinjection, and their lungs examined both macro- andmicroscopically (Fig. 7a). Tumour masses show high levels ofnuclear and cytoplasmic Ccnd1 and phosphorylated Pxn at serine83 (Fig. 7b). Downregulation of Ccnd1 by RNA interferencedrastically reduced R3327-50A-dependent metastases. Strikingly,the presence of a phosphomimetic S83,178E Pxn allele rescuedthe metastatic potential of these cells (Fig. 7a,c). This effect wasnot due to changes in the proliferative potential of the cells.Ccnd1 is important to maintain a high-proliferation rate intransformed cell lines and, as expected, downregulating Ccnd1reduced the proliferative capacity of R3327-50A cells to a half(Fig. 7d). Yet, a similar reduction in proliferation was observed incells expressing the phosphomimetic allele of Pxn. Therefore, therescue of the metastatic potential cannot be attributed to changesin proliferation. We then propose that phosphorylation of Pxn byCcnd1 �Cdk4 is a new mechanism whereby Ccnd1 promotesmetastasis.

DiscussionThe best studied role of the Ccnd1 �Cdk4 complex is as aregulator of transcription in the nucleus. However, some studieshave also proposed a cytoplasmic function for the complex10,12,13.The accumulation of Ccnd1 �Cdk4 outside the nucleus wasinitially described as a mechanism to arrest cell proliferation. Forinstance, oncogenic Ras induces re-localization of Ccnd1 �Cdk4

in the cytoplasm and promotes proliferation arrest in primaryhuman keratinocytes, probably as a mechanism of defence againstRas-driven neoplasia27. Another example is the regulation ofproliferation by tight junctions through the sequestration ofCcnd1 �Cdk4 in the membrane of MDCK-epithelial cells28. Here,we show that the localization of Ccnd1 �Cdk4 in the membrane offibroblasts and tumour cells has an active role in the induction ofcell migration and invasion through the phosphorylation of Pxn.Our findings do not exclude the re-localization of Ccnd1 �Cdk4as a mechanism of proliferation control, but indicate the existenceof cytoplasmic substrates of Ccnd1 �Cdk4; moreover, we showthat Ccnd1 �Cdk4 is involved in the regulation of cell-matrixadhesion and cell migration through the phosphorylation of asubpopulation of cytoplasmic Pxn molecules. Ccnd1 binds to andco-localizes with Pxn in the cell membrane, its removal orinhibition leads to decreased levels of Pxn phosphorylation, and itphosphorylates Pxn in vitro. In particular, Ccnd1-dependentphosphorylation of Pxn at serine 83 in vivo is required andirreplaceable in the regulation of cell spreading and invasion. Bycontrast, the effect on cell spreading of serine-178phosphorylation, although required, does not rely entirely onCcnd1. In this respect, c-Jun N-terminal kinase (JNK) kinase hasbeen shown to promote cell spreading and migration in epithelialcells through the phosphorylation of Pxn at serine 178 (refs22,33). Also, we have observed that Ccnd1 �Cdk4 in vitrophosphorylates Pxn at S244. Although phosphorylation at thissite has not been associated with cell adhesion and migration, wecannot rule out the possibility that the phosphorylation at S244by Ccnd1?Cdk4 could play a role in vivo in those processes.

Phosphorylation of Pxn at serine 83 by Erk promotes cellspreading and migration23. However, Erk and Ccnd1 �Cdk4 havea discordant effect on cell spreading modulation. Hepatocytegrowth factor (HGF)-induced Erk activation promotes cellspreading in epithelial cells,23 whereas knockdown ofCcnd1 �Cdk4 enhances cell spreading in mouse fibroblasts. Thisdiscrepancy could be related to the different localization of S83-phosphorylated Pxn. We have observed S83-phospho-Pxn both atthe cell membrane and in FAs, but we have detected co-localization with Ccnd1 only at the membrane. By contrast,localization of Erk to FAs has been described23. Note that Ccnd1binds to the C-terminal region (LIM domains) of Pxn (Fig. 1d),and that LIM domains are required for efficient targeting of Pxnto FAs34. Then, both interactions could be mutually exclusive.Conceivably, Pxn phosphorylation by Erk at FAs may lead tomore efficient cell spreading while Pxn phosphorylation by Ccnd1at the cell membrane may lead to an opposite effect. This is not afar-out possibility. For instance, Y31/118-phosphorylated Pxn ispresent at different locations promoting different effects on celladhesion26. The tyrosine kinases FAK and Brk1 phosphorylatePxn at FAs and lamellipodia, respectively, and both promote cellinvasion. However, phosphorylation of Pxn by FAK is importantfor FA turnover and cell adhesion whereas phosphorylation ofPxn by Brk1 does not alter cell adhesion26.

Rac1 function is essential for membrane ruffling and protrusiveactivity of cells35. Similar to Ccnd1� /� cells, Rac1� /�

fibroblasts show a more spread morphology than wild-type cellsand are impaired in migration36. In this work, we propose thatCcnd1 �Cdk4 induces Rac1 activation through thephosphorylation of Pxn at serine 83 in the cell membrane. Weshow that Ccnd1-deficient fibroblasts not only have a reductionin Pxn phosphorylated at serine 83 but also in Rac1-GTP levels,and exhibit a decrease in membrane ruffling after seeding infibronectin. In addition, the expression of a hyperactive allele ofRac1 rescues the invasion and spreading phenotypes observed inCcnd1-deficient cells to the same degree as a phosphomimeticallele of Pxn. Importantly, Ccnd1 �Cdk4 only induces Rac1




activation in the presence of a wild-type allele of Pxn but not inthe presence of an S83,178A non-phosphorylatable Pxn mutant.This result raises the question as to how S83-phosphorylated Pxnactivates Rac1. In FAs, phosphorylation of Pxn at serine 83 by Erkenhances the interaction of Pxn with FAK, which consequentlypromotes the phosphorylation of Pxn at tyrosines 31 and 118 (ref. 23).In turn, phosphorylation of Pxn at these tyrosines leads tothe recruitment of a GEF factor (Dock180) that induces Rac1activity and cell migration20. This mechanism cannot be applied

in our case because Ccnd1� /� fibroblasts show an increase inthe amount of Pxn phosphorylated at tyrosine 118 (ref. 8), whilewe have observed lower levels of activated Rac1 in Ccnd1� /�

cells. Then there must be alternative pathways involving otherGEFs recruited by Pxn (such as b-PIX and Vav2) that couldmediate Ccnd1-dependent activation of Rac1 (refs 20,37).

At first, cyclins, Cdks and Cdk-inhibitors were exclusivelyviewed as nuclear proteins involved in cell cycle transitions.However, emerging data demonstrate that these cell cycle

Control

shD1

Pxnwt

Pxnwt

shD1

PxnS83Es178E

PxnS83ES178EshD1

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

Num

ber

of c

ells

(10

,000

×)

a

Vector

Pxnwt

PxnS83ES178E

Control ControlshD1 shD1

Lungs

** ** ns

d

Ccnd1 Ki67

10×

Nodule40×

B

**

****

**

ns

Control

shD1

Pxnwt

Pxnwt

shD1

PxnS83ES178EshD1

0.0

1.0

2.0

3.0

4.0

5.0

Met

asta

ses

per

mm

2

R3327-5′A cellsR3327-5′A cells

PxnS83ES178E

Normal40×

PhoS83 Pxn

Hematoxylin–eosin staining

b

c

Figure 7 | Phosphomimetic Pxn rescues the low metastatic potential of Ccnd1-deficient cells. (a) For metastasis assays, 5� 105 co-infected R3327-50A

cells as in Fig. 3 were inoculated in nude mice (four animals per condition) by retroorbital intravenous injection, and the animals were euthanized fifteen

days later. Lungs were recovered, fixed in Bouin’s solution (3 mm bar), and a sample was included in paraffin for hematoxylin–eosin staining (200mm bar).

(b) A piece of biopsy was fixed with formaldehyde at 4%, included in paraffin and processed by immunohistochemistry (IHC) to detect Ccnd1,

phosphorylated Pxn and Ki67 (25mm bar). (c) Metastatic capacity was evaluated as number of metastases per mm2, and expressed as mean±s.e.m.

(n¼4) with significance values determined by one way ANOVA and Tukey-HSD post-test (**Po0.01; ns no significant). (d) For proliferation assays,

2� 104 co-infected cells were seeded and grown in DMEM 10% serum at 37 �C, 5% CO2. After three days, cell number was determined and plotted.

The experiment was repeated five times and data are expressed as mean±s.e.m. with significance values determined by one way ANOVA and Tukey-HSD

post-test (**Po0.01).




regulators also operate in the cytoplasm regulating cell functionsindependently of their cell cycle role. One of these functions is cellmigration, which requires a tight coordination between Rho andRac1 activities35. Rho activity has to be reduced, whereas Rac1activity has to be increased in the leading edge of the cell. TheCip/Kip inhibitor p27 induces migration through the binding andinhibition of RhoA activity in the cytoplasm38. Also, the INK4Cdk-inhibitor p16 promotes cell migration in hepatocellularcarcinoma cells through the activation of Rac1, although it hasalso been described that p16 plays a negative role on migration inother tumours39. Finally, the Ccnd1 �Cdk4 complex promotes cellmigration through different cytoplasmic mechanisms. Pestell andcollaborators have previously shown that Ccnd1 regulates cellmigration by transcriptional repression of RhoA effectors8, andby modulating p27 levels, which inhibits RhoA signalling in thecytoplasm9. In this work we show that Ccnd1 �Cdk4 promotescell migration through Rac1 activation in the cytoplasm.Therefore, Ccnd1 �Cdk4 seems to favour cell migration throughthe regulation of two types of Rho small GTPases, RhoA andRac1, in an opposite way. Altogether, it seems that cell cycleregulators show a dual functionality regulating proliferation inthe nucleus and migration in the cytoplasm. Perhaps, this featureis relevant in cell fate decision during development. Ccnd1accumulates in the cytoplasm of post-mitotic neurons40 andduring skin differentiation41 where it could act through Pxn toregulate migratory events throughout development.

The entrance of a malignant tumour in a metastatic process isoften intractable and is a major concern in cancer therapeutics.Metastatic cells modify their ability to adhere to the extracellularmatrix and acquire proficient motility. Ccnd1-dependent activityis frequently increased during tumour growth and metastasis3–5,42.Its importance in these processes has been attributed not only toits role in the control of cell proliferation but also to the regulation ofcell-matrix adhesion and cell migration6. Here, we havedemonstrated that the inhibitory effect on the metastatic potentialof cells upon downregulation of Ccnd1 can be reverted by theexpression of a phosphomimetic version of Pxn. This agrees with thereported requirement of phospho-Pxn for the induction ofmetastases in different tumour models43–46. Our data place Pxndownstream of Ccnd1, as a phosphorylation target, in a pathwayregulating cell spreading, invasion and metastasis. This implies thataccumulation of Ccnd1 in the cytoplasm would exert a keyactivating role on the metastatic potential, a correlation that has beenobserved in prostate tumours47 and in cancer cell lines capable ofundergoing metastasis in in vivo models48. At present, the cyclinD �Cdk4,6 complexes are considered relevant targets for cancertherapy and at least three different specific inhibitors of Cdk4/6 arebeing used in different clinical trials49. For instance, Palbociclib hasalready been approved for the treatment of oestrogen receptor-positive, HER2-negative metastatic breast cancer. Our data reinforcethe importance of these inhibitors in cancer therapy suggesting thatthe inactivation of Cyclin D �Cdk4,6 not only should producetumour regression, as expected from its role in cell proliferation, butshould restrict tumour spreading and metastasis as well.

MethodsCell culture. CCND1� /� and CCND1þ /þ fibroblasts and R3327-50 rat tumourcells were kindly provided by P. Sicinski and M. Hendrix respectively (growthconditions, morphology and chromosome number were authenticated followingprovider’s instructions). HEK293T cells were obtained from the American TypeCulture Collection. Primary MEFs were isolated from E14.5 wild type or CCND1� /�

embryos. All the cells were routinely tested for mycoplasma by PCR and main-tained mycoplasma free. Our cell lines are not listed in ICLAC database. Cells weremaintained at 37 �C in a 5% CO2 incubator, and grown in Dulbecco’s modifiedEagle’s medium (DMEM) supplemented with 10% FBS, 100 mg ml� 1 penicillin/streptomycin and 2 mM glutamine. Transient transfection of vectors wasperformed with Lipofectamine 2000 (Invitrogen) according to manufacturer’sinstructions. For lentivirus production, HEK293T cells were transfected with

lentiviral expression vectors, envelope plasmid pVSV.G, and packaging plasmidpHR’82DR at a ratio 2:1:1. Stable knockdown of Ccnd1 was carried out by lentiviralinfection of R3327-50A cells and selection in 5 mg ml� 1 of puromycin.

Expression vectors. Both human CCND1 wild type and the inactive allele K112Ewere fused to three copies of the FLAG epitope under the CMV promoter inpcDNA3. Also, the human CCND1 was used to obtain an N-terminal 3�HAfusion under the UBI promoter in a lentiviral vector derived from pDSL(Invitrogen). Mouse Pxn (IMAGE ID 5309957 clone from Source BioScience) wasused to obtain an N-terminal GST fusion in pGEX-KG (Clontech) or anN-terminal 3�HA fusion under the UBI promoter in pDSL-derivative vector. GST-Pxn contains full length Pxn, while GST-Pxn-Ct contains only the four LIM domainsof the protein (aa337-591). Both Pxn wild type and the S178E allele inserted into themammalian GFP-tagged expression vector pEGFP-C1 (BD Biosciences) were kindlyprovided by J. Yamauchi50. Standard PCR-mediated site-directed mutagenesis wasused to obtain the non-phosphorylatable and phosphomimetic mutants of Pxn usingthe following primers: GCCCCGCAGCGAGTCACCTCCAG and GCCCATCTCTCCCTGGTTCACAGT for S244A; GCGCCACTGCCCGTGTACAGCTC andCGGAGGCTGCTGGTGAGCGT for S83A; GCGCCCCTTTATGGCATCCCAGAand CAGGGCTCCAGGCAAGGGGGG for S178A; CCGGTGTACAGCTCCAGTGCTAA and TAGTGGCTCCGGAGGCTGCTGGTGAGCGT for S83E;GAGCCCCTTTATGGCATCCCAGA and GAGGGCTCCAGGCAAGGGG for S178E.The single, double and triple mutants were done in pGEX-KG (GST-Paxillin) and thentransferred to pcDNA3 and pDSL vectors. The GST-pRb1 (aa379–928) fusion was a giftfrom N. Agell. The pCEFL-AU5-Rac1(Q61L) plasmid was obtained from P. Crespo viaX. Bustelo. For RNA interference the CCND1 MISSION shRNA TRCN0000026883cloned in a pLKO.1-puro was obtained from Sigma. The YFP-PBD vector was obtainedfrom J. Swanson via Addgene.

Immunofluorescence. Briefly, cells were quickly washed in PBS and fixed in 4%paraformaldehyde for 15 min at room temperature. Fixed MEFs and R3327-50Acells were permeabilized with 0.2% Triton-X-100 for 3 min at room temperature,and blocked with 3% BSA. In Fig. 5b low permeabilization conditions were used(0.02% Triton-X-100). Primary antibodies at a working dilution 1:200 were com-bined with adequate Alexa488 and/or Alexa594-labelled secondary antibodies(Molecular Probes) in PBS with 0.3% BSA. Nuclei were stained with Hoechst(Sigma). Images were acquired using 40X and 60X objectives in an OlympusFV1000 confocal system. Primary antibodies used in IF were the following: anti-Ccnd1 (rabbit monoclonal clone EP12, Dako #M3642/ mouse monoclonal clone72-13 G, Santa Cruz #sc450), anti-Pxn (S83) phospho-specific (polyclonal#PP1341, ECM Biosciences), anti-TFR (monoclonal H68.4, Invitrogen #13–6800),anti-Pxn (monoclonal 349, BD Biosciences #610051), anti-Rac1 (mouse mono-clonal, clone 102, BD #610650) and anti-HA (rat monoclonal 3F10, Roche#11867431001, 1:1000). For detection of YFP-PBD we have used two different anti-GFPs antibodies (mouse monoclonal 3E6, Invitrogen #A11120, 1:200; rabbit AlexaFluor 488 conjugate, Invitrogen #A21311, 1:400). The quantification of cells withPxn and Rac1 accumulation in ruffles was carried out with Image J software (seeSupplementary Fig. 6).

Immunohistochemistry. Lung tissue samples were fixed with paraformaldehyde.Blocks were sectioned at a thickness of 3 mm and dried for 1 h at 65 �C, before beingdewaxed in xylene and rehydrated through a graded ethanol series, then washedwith PBS. Antigen retrieval was performed by heat treatment in a pressure cookerfor 2 min in EDTA (pH 8.9). Before staining the sections, endogenous peroxidasewas blocked. The antibodies used were anti-Ccnd1 (rabbit monoclonal clone EP12,Dako #M3642, 1:400), anti-Pxn S83 phospho-specific (polyclonal #PP1341, ECMBiosciences, 1:100), and anti-K67 (rabbit monoclonal clone SP6, Abcam #ab16667,1:250). After incubation, the reaction was visualized with the EnVision DetectionKit (Dako), using diaminobenzidine chromogen as a substrate.

Cell fractionation and immunoprecipitation. Protein fractionation was per-formed with the Subcellular Protein Fractionation kit for cultured cells (ThermoScientific-Pierce; 78840). Soluble fraction corresponds to a mixture of cytosolic andnuclear soluble fractions described in the supplier’s instructions.

IP of endogenous Ccnd1 was carried out in MEFs with a rabbit polyclonalantibody (#06–137, Upstate, 5 mg). IP of endogenous Pxn was done in Ccnd1� /�

and Ccnd1þ /þ immortalized MEFs with an anti-Pxn rabbit polyclonal antibody(H-114, Santa Cruz #sc-5574, 2 mg). Briefly, cleared cell extracts in lysis buffer(10 mM Tris–HCl, pH 7.5, 50 mM NaCl, 1% Triton X-100, 3 mM MgCl2, 300 mMSucrose, and protease and phosphatase inhibitors) were immunoprecipitated withProtein A linked to magnetic beads (Dynabeads, Invitrogen). An anti-FLAG(#F7425, Sigma) was used as a mock control. The same lysis buffer was used forIP of GFP-tagged Pxn in transfected R3327-50A cells. Cleared extracts wereimmunoprecipitated with a mixture of anti-GFP monoclonal antibodies (Roche).

Immunoblotting. For immunoblot, protein samples were resolved by SDS-PAGE,transferred to PVDF membranes (Millipore), and incubated with primary anti-bodies anti-Ccnd1 (monoclonal DCS-6, BD Pharmigen #556470, 1:500), anti-Cdk4




(polyclonal C-22, #sc-260, 1:250), anti-Pxn (monoclonal 349, BD transduction lab.#610051, 1:1000), anti-GFP (monoclonals 7.1 and 13.1, Roche #11814460001,1:2000), anti-Flag (monoclonal M2, Sigma #F3165, 1:2000), anti-Pxn (ser-83)phospho-specific (polyclonal #PP1341, ECM Biosciences, 1:500), anti-HA (ratmonoclonal 3F10, Roche #11867431001, 1:2000), anti-PCNA (monoclonal PC10,Abcam #ab29, 1:1000), anti-TFR (monoclonal H68.4, Invitrogen #13–6800,1:1,000), anti-GAPDH-Peroxidase (monoclonal clone 71.1, Sigma #G9295,1:40000), anti-ERK1/2 (monoclonal MK12, Merck #05–1152, 1:2000), anti-phos-pho-ERK1/2 (Thr202/Tyr204, polyclonal, Cell Signalling #9101, 1:800), anti-GST(goat polyclonal, Amersham #27–4577, 1:2000) and anti-tubulin (monoclonal B-5-1-2 Sigma #T5168, 1:10000). Appropriate peroxidase-linked secondary antibodies(GE Healthcare UK Ltd) were detected using the chemiluminescent HRP substrateImmobilon Western (Millipore). Chemiluminescence was recorded with a Che-miDoc-MP imaging system (BioRad). Uncropped scans of the most importantwestern blots are supplied in the Supplementary Figs 10–13.

GST pull-down assay. Flag tagged Ccnd1 was transcribed with T7 RNA poly-merase (New England Biolabs) and translated in Rabbit Reticulocyte Lysate System(Promega). For the pull-down assay, 400 ng of GST or GST-Pxn purified fromE. coli were immobilized on Glutathione-Sepharose 4B beads and incubated withFlag-Ccnd1 in binding buffer (20 mM HEPES-KOH, pH 7.5, 150 mM KCl, 5 mMMgCl2, 0.5 mM EDTA, 0.1% NP-40, 1 mM DTT, 1 mM PMSF, 10% glycerol,protease and phosphatase inhibitors) for 30 min at room temperature. After fourwashes with the same buffer, the samples were analyzed by SDS-PAGE.

Kinase assay. Kinase reaction and purification of GST-Pxn substrates from E. coliwere done as previously described12. Briefly, 0.2 mg substrate (either GST-Pxn orGST-pRb1) was mixed with 1.5 ml of active Ccnd1-Cdk4 complex purified frombaculovirus (Sigma C0620), 10 mM ATP, 7 mCi of g-32P-ATP (PerkinElmer,3000 Ci/mmol), and either DMSO or 2 mM of the Cdk4/6-inhibitor Palbociclib(Selleckchem, S1116) in 20 ml of kinase buffer (50 mM Tris–HCl pH 7.5, 10 mMMgCl2, 0.5 mM DTT, 1 mM EGTA and 2.5 mM b-glycerophosphate). This mixturewas incubated for 20 min at 30 �C, then boiled in 2� Laemmli buffer, andseparated by electrophoresis. Phosphorylated proteins were visualized byautoradiography of the dried slab gels. For substrate purification, E. coli cellstransformed with GST fusions were grown to saturation overnight, diluted 1:10 inLB broth, and incubated at 37 �C for 2 h. GST-fusion proteins were induced byaddition of IPTG for 4 h at 30 �C, after which cells were recovered by centrifugationat 4 �C and lysed on ice by sonication in 1 ml of lysis buffer (50 mM HEPES, pH7.5,150 mM NaCl, 1 mM EDTA, 1 mM DTT, 10% glycerol, 0.5% Triton X-100).Cleared lysates were mixed with Glutathione-Sepharose 4B (GE, Healthcare) andincubated for 2 h at 4 �C. Beads were washed three times with lysis buffer and twicewith kinase buffer, and the GST-fusion proteins were released by incubation with2 mM reduced glutathione (Sigma). The concentration and purity of substrateswere estimated by comparison to protein standards stained with Coomassie blue.

In-gel digestion. Full length Pxn fused to GST was used in an in vitro kinase assaywith Ccnd1-Cdk4 in the presence of ATP or in the absence of ATP as a control.Samples were subsequently subjected to SDS-PAGE. The gels were stained withCoomassie Brilliant Blue G-250 colloidal (EZBlue Gel Staining Reagent, Sigma).After washing with water, protein bands of interest were prepared and submitted toFundacio Institut d’Investigacio Biomedica de Bellvitge (IDIBELL) ProteomicsService (Barcelona) for analysis of the Chymotryptic peptide molecular masses byliquid chromatography-mass spectrometry. Gel slices were manually cut andproteomic service suggested for each band-assay to recover three slices (low 1,middle 2, and high 3 mobility) as phosphorylation could alter band mobility.

Briefly, gel bands were washed with water, ammonium bicarbonate (50 mM)and 50% acetonitrile. Next, samples were reduced by incubation with dithiothreitol(10 mM) at 60 �C for 45 min and alkylated with iodoacetamide (50 mM) for 30 min,in the dark. Finally, proteins were digested with chymotrypsin (5 ng ml� 1) at 25 �Covernight (Trypsin gold, Promega). Digestion was stopped by addition of 5%formic acid and peptides extracted twice with 70% acetonitrile and 5% formic acid(10 min sonication). Peptide extracts were evaporated to dryness, resuspended with2% acetonitrile 0.1% formic acid and analyzed by nano-HPLC-MSMS.

LC-MSMS and database searching. Peptides were analyzed using an Easy-nanoLCII (Proxeon, Denmark) coupled to an Amazon ETD Ion trap (BrukerDaltonics). Peptides were first trapped on an Easy column TM C18 (2 cm, 5 mm,100mm ID, Thermo Scientific), and then separated using an analytical C18nanocapillary column (75 mM ID, 15 cm, AcclaimPepMap 100 Thermo Scientific).The chromatography gradient was achieved by increasing percentage of buffer Bfrom 0–35% at a flow rate of 300 nl min� 1 over 40 min (A: 0.1% formic acid, B:0.1% formic acid, 100% acetonitrile). Eluted peptides were then introduced intoAmazon ETD Ion trap by electrospray ionization in the Captive-Spray ion Source(Bruker Daltonics) with an applied voltage of 1,500 V and N2 as drying gas.Peptide masses (400–1,400 m/z) were analyzed at full scan at enhanced resolution(range 50–3,000 m/z and speed 8.100 m/z sec) and 10 most intense peptides wereselected and fragmented in a three-dimensional ion trap using both collisioninduced dissociation and electron transfer dissociation fragmentation, using as a

collision gas hellium and methane, respectively. Data was generated with DataAnalyst 4.1 software (Bruker Daltonics). MS and MS/MS data were analyzed withProtein Scape 3.1.2 software (Bruker Daltonics) using Mascot 2.4.0 (MatrixScience) as the search engine and SwissProt database (2015-03, 547,964 sequences;195,174,196 residues). The specific parameters for protein sequence databasesearching included taxonomy Mus musculus (16,711 sequences), serine, threonineand tyrosine phosphorylation and methionine oxidation as variable modificationsand cysteine carbamylation as a fixed modification. Other search parameters were:chymotrypsin digestion with two missed cleavages, charge states þ 1, þ 2 þ 3 forprecursor ion, mass error of 0.6 Da for precursor ion and for fragment ions. For theidentification a significance threshold peptide decoy (Mascot) of Po0.05 was setand only peptides with a minimum Mascot score of 25 and proteins with aminimum Mascot score of 35 were considered. Spectra from phosphorylatedpeptides were also manually examined.

Rac1 pull-down assay. The assays were performed by using PAK1 PBD agarose-beads (Cell Biolabs, STA-411) according to the manufacturer’s instructions. Celllysates were obtained from one 100 mm plate from fibroblasts or tumour cells. Thelysis buffer used was 50 mM Tris pH 7.5, 200 mM NaCl, 2.5 mM MgCl2, 2.5 mMDTT, 1% Triton and protease and phosphatase inhibitors. Lysates (0.6 ml) wereincubated with 10mg of PAK1 beads during 30 min at 4 �C and, after severalwashes, agarose beads were resuspended in 2� Laemmli buffer. Samples wereseparated by SDS-PAGE, transferred to PVDF membranes, and immunoblotted.

Cell spreading assay. Fibroblasts or R3327-50A cells were co-transfected withGFP-Pxn wild type or any of the mutated alleles together with HA-Ccnd1 or anempty vector. Petri dishes were coated overnight at 4 �C with a 5 mg ml� 1 solutionof fibronectin (Invitrogen) in PBS. Forty-eight hours after transfection, cells weretrypsinized and seeded in serum-free medium in fibronectin-coated 35-mm wellplates. Fibroblasts were incubated for thirty minutes and R3327-50A cells for 1 h,and then cells were fixed with 2% paraformaldehyde for 10 min on ice, imaged, andcells that had spread counted (nZ150 total green cells evaluated in each inde-pendent experiment). Round and bright cells were considered to be unspread.

Cell invasion assay. We performed cell invasion assays with R3327-50A cells aspreviously described51. Briefly, 6.5-mm filters of 8.0 pore size (Transwell, Corning)were coated with Matrigel (reduced-factors, BD Biosciences) in the upper side.Then, infected cells (5� 104) were seeded in the bottom side of the filter for fourhours to allow their attachment. Afterwards, filters were loaded with DMEM 10%serum and incubated in 24-well plates containing serum-free medium for 24 h.Under these conditions, some cells migrate from the bottom to the upper side ofthe filter invading the Matrigel. The remaining cells at the bottom of the filter wereremoved, and Matrigel-embedded cells were fixed and stained with Hoescht. Wehave counted all cells in the filter extension with the Image J software.

Metastasis assay. The procedure performed in this study followed the NationalInstitutes of Health Guidelines for the Care and Use of Laboratory Animals, and isaccording to the Article 17.2 of the Law 5/1995 and the Article 33.a of Decree 214/1997 of 30 July, which regulate the use of animals for experimental and otherscientific purposes (Catalan Government), and was certified by the Ethics Com-mittee on Animal Experimentation from the University of Lleida (CEEA 03-03/13).Immunodeficient female SCID hr/hr mice (12-week-old; 20–25 g) were maintainedin specific pathogen free conditions, and were inoculated with 5� 105 co-infectedcells by retroorbital intravenous injection. Animals were euthanized fifteen daysafterwards. Lungs were recovered and fixed in Bouin’s solution; a sample wasincluded in paraffin for hematoxylin–eosin staining.

Statistical analyses. The investigators were blinded when assessing the outcomefor experiments of spreading (Fig. 2b), invasion (Fig. 3a) and immuno-localization(Fig. 6b). Comparisons among groups were made by one way ANOVA and Tukey-HSD post-test or with two-tailed t-test allowing unequal variance (*Po0.05,**Po0.01, ns no significant). Throughout the paper, error bars indicate s.e.m.except for figures where we only have two independent experiments, such as Figs 1fand 6b,f (mean±s.d.). In Fig. 5d and Supplementary Figs 6B, 9B,D error bars referto the confidence limits for a proportion.

For animal studies we have used the Ene 3.0: Program to calculate sample size.This software was developed by the Department of Applied Statistics ofAutonomous University of Barcelona and is distributed by GlaxoSmithKline. Nospecific method of randomization was used but the group allocation was donerandomly. The program suggested a minimum of 18 mice distributed among sixgroups (at least three animals per group) to get a 0.8 power to detect differences inthe null hypothesis H0 (the mean of six groups they are equal) by one factorANOVA test for independent samples, considering the significance level is 0.05.We have used 24 animals (four animals per group).




References1. Sherr, C. J. & Roberts, J. M. Living with or without cyclins and cyclin-

dependent kinases. Genes Dev. 18, 2699–2711 (2004).2. Bienvenu, F. et al. Transcriptional role of cyclin D1 in development revealed by

a genetic-proteomic screen. Nature 463, 374–378 (2010).3. Drobnjak, M., Osman, I., Scher, H. I., Fazzari, M. & Cordon-Cardo, C.

Overexpression of cyclin D1 is associated with metastatic prostate cancer tobone. Clin. Cancer Res. 6, 1891–1895 (2000).

4. Huang, H., Hu, Y., Li, N. & Zhu, Y. Inhibition of tumor growth and metastasisby non-small cell lung cancer cells transfected with cyclin D1-targeted siRNA.Oligonucleotides 19, 151–162 (2009).

5. Zheng, L. et al. microRNA-9 suppresses the proliferation, invasion andmetastasis of gastric cancer cells through targeting cyclin D1 and Ets1. PLoSONE 8, e55719 (2013).

6. Velasco-Velazquez, M. A. et al. Examining the role of cyclin D1 in breastcancer. Future Oncol. 7, 753–765 (2011).

7. Neumeister, P. et al. Cyclin D1 governs adhesion and motility of macrophages.Mol. Biol. Cell 14, 2005–2015 (2003).

8. Li, Z. et al. Cyclin D1 regulates cellular migration through the inhibitionof thrombospondin 1 and ROCK signaling. Mol. Cell. Biol. 26, 4240–4256(2006).

9. Li, Z. et al. Cyclin D1 induction of cellular migration requires p27(KIP1).Cancer Res. 66, 9986–9994 (2006).

10. Zhong, Z. et al. Cyclin D1/cyclin-dependent kinase 4 interacts with filamin Aand affects the migration and invasion potential of breast cancer cells. CancerRes. 70, 2105–2114 (2010).

11. Alhaja, E. et al. Anti-migratory and anti-angiogenic effect of p16: a novellocalization at membrane ruffles and lamellipodia in endothelial cells.Angiogenesis 7, 323–333 (2004).

12. Fernandez, R. M. H., Ruiz-Miro, M., Dolcet, X., Aldea, M. & Garı, E. Cyclin D1interacts and collaborates with Ral GTPases enhancing cell detachment andmotility. Oncogene 30, 1936–1946 (2011).

13. Meng, H. et al. PACSIN 2 represses cellular migration through directassociation with cyclin D1 but not its alternate splice form cyclin D1b. CellCycle 10, 73–81 (2011).

14. Li, Z. et al. Cyclin D1 integrates estrogen-mediated DNA damage repairsignaling. Cancer Res. 74, 3959–3970 (2014).

15. Glenney, J. R. & Zokas, L. Novel tyrosine kinase substrates from Rous sarcomavirus-transformed cells are present in the membrane skeleton. J. Cell Biol. 108,2401–2408 (1989).

16. Turner, C. E., Glenney, J. R. & Burridge, K. Paxillin: a new vinculin-bindingprotein present in focal adhesions. J. Cell Biol. 111, 1059–1068 (1990).

17. Geiger, B., Bershadsky, A., Pankov, R. & Yamada, K. M. Transmembraneextracellular matrix-cytoskeleton crosstalk. Nat. Rev. Mol. Cell Biol. 2, 793–805(2001).

18. Parsons, J. T., Horwitz, A. R. & Schwartz, M. A. Cell adhesion: integratingcytoskeletal dynamics and cellular tension. Nat. Rev. Mol. Cell Biol. 11, 633–643(2010).

19. Hagel, M. et al. The adaptor protein paxillin is essential for normaldevelopment in the mouse and is a critical transducer of fibronectin signaling.Mol. Cell. Biol. 22, 901–915 (2002).

20. Deakin, N. O. & Turner, C. E. Paxillin comes of age. J. Cell Sci. 121, 2435–2444(2008).

21. Nayal, A. et al. Paxillin phosphorylation at Ser273 localizes a GIT1-PIX-PAKcomplex and regulates adhesion and protrusion dynamics. J. Cell Biol. 173,587–589 (2006).

22. Huang, C., Rajfur, Z., Borchers, C., Schaller, M. D. & Jacobson, K. JNKphosphorylates paxillin and regulates cell migration. Nature 424, 219–223(2003).

23. Ishibe, S., Joly, D., Liu, Z.-X. & Cantley, L. G. Paxillin serves as an ERK-regulated scaffold for coordinating FAK and Rac activation in epithelialmorphogenesis. Mol. Cell 16, 257–267 (2004).

24. Miyamoto, Y. et al. Cdk5 regulates differentiation of oligodendrocyte precursorcells through the direct phosphorylation of paxillin. J. Cell Sci. 120, 4355–4366(2007).

25. Matafora, V., Paris, S., Dariozzi, S. & de Curtis, I. Molecular mechanismsregulating the subcellular localization of p95-APP1 between the endosomalrecycling compartment and sites of actin organization at the cell surface. J. CellSci. 114, 4509–4520 (2001).

26. Chen, H.-Y. et al. Brk activates rac1 and promotes cell migration and invasionby phosphorylating paxillin. Mol. Cell. Biol. 24, 10558–10572 (2004).

27. Lazarov, M. et al. CDK4 coexpression with Ras generates malignant humanepidermal tumorigenesis. Nat. Med. 8, 1105–1114 (2002).

28. Balda, M. S., Garrett, M. D. & Matter, K. The ZO-1-associated Y-box factorZONAB regulates epithelial cell proliferation and cell density. J. Cell Biol. 160,423–432 (2003).

29. Luo, J. et al. Heterogeneous expression of invasive and metastatic properties ina prostate tumor model. Pathol. Oncol. Res. 3, 264–271 (1997).

30. Ridley, A. J., Paterson, H. F., Johnston, C. L., Diekmann, D. & Hall, A. Thesmall GTP-binding protein rac regulates growth factor-induced membraneruffling. Cell 70, 401–410 (1992).

31. Bretscher, M. S. & Aguado-Velasco, C. EGF induces recycling membrane toform ruffles. Curr. Biol. 8, 721–S4 (1998).

32. Hoppe, A. D. & Swanson, J. A. Cdc42, Rac1 and Rac2 display distinct patternsof activation during phagocytosis. Mol. Biol. Cell 15, 1895–1903 (2004).

33. Huang, Z., Yan, D.-P. & Ge, B.-X. JNK regulates cell migration throughpromotion of tyrosine phosphorylation of paxillin. Cell Signal. 20, 2002–2012(2008).

34. Brown, M. C., Perrotta, J. A. & Turner, C. E. Identification of LIM3 as theprincipal determinant of paxillin focal adhesion localization andcharacterization of a novel motif on paxillin directing vinculin and focaladhesion kinase binding. J. Cell Biol. 135, 1109–1123 (1996).

35. Iden, S. & Collard, J. G. Crosstalk between small GTPases and polarity proteinsin cell polarization. Nat. Rev. Mol. Cell Biol. 9, 846–859 (2008).

36. Steffen, A. et al. Rac function is crucial for cell migration but is not required forspreading and focal adhesion formation. J. Cell Sci. 126, 4572–4588 (2013).

37. Jones, M. C., Machida, K., Mayer, B. J. & Turner, C. E. Paxillin kinase linker(PKL) regulates Vav2 signaling during cell spreading and migration. Mol. Biol.Cell 24, 1882–1894 (2013).

38. Besson, A., Gurian-West, M., Schmidt, A., Hall, A. & Roberts, J. M. p27Kip1modulates cell migration through the regulation of RhoA activation. Genes Dev.18, 862–876 (2004).

39. Chen, Y.-W. et al. p16 Stimulates CDC42-dependent migration ofhepatocellular carcinoma cells. PLoS ONE 8, e69389 (2013).

40. Sumrejkanchanakij, P., Tamamori-Adachi, M., Matsunaga, Y., Eto, K. & Ikeda,M.-A. Role of cyclin D1 cytoplasmic sequestration in the survival of postmitoticneurons. Oncogene 22, 8723–8730 (2003).

41. Fernandez-Hernandez, R. et al. Cyclin D1 localizes in the cytoplasm ofkeratinocytes during skin differentiation and regulates cell-matrix adhesion.Cell Cycle 12, 2510–2517 (2013).

42. Beroukhim, R. et al. The landscape of somatic copy-number alteration acrosshuman cancers. Nature 463, 899–905 (2010).

43. Deakin, N. O., Pignatelli, J. & Turner, C. E. Diverse roles for the paxillin familyof proteins in cancer. Genes Cancer 3, 362–370 (2012).

44. Chen, D.-L. et al. Overexpression of paxillin induced by miR-137 suppressionpromotes tumor progression and metastasis in colorectal cancer. Carcinogenesis34, 803–811 (2013).

45. Lu, W. et al. The roles of Wnt5a, JNK and paxillin in the occurrenceof metastasis of pancreatic adenocarcinoma. Int. J. Clin. Oncol. 19, 1011–1019(2013).

46. Chen, J. & Gallo, K. A. MLK3 regulates paxillin phosphorylation in chemokine-mediated breast cancer cell migration and invasion to drive metastasis. CancerRes. 72, 4130–4140 (2012).

47. Fleischmann, A. et al. High-level cytoplasmic cyclin D1 expression in lymphnode metastases from prostate cancer independently predicts early biochemicalfailure and death in surgically treated patients. Histopathology 58, 781–789(2011).

48. Alao, J. P. et al. The cyclin D1 proto-oncogene is sequestered in the cytoplasmof mammalian cancer cell lines. Mol. Cancer 5, 7 (2006).

49. VanArsdale, T., Boshoff, C., Arndt, K. T. & Abraham, R. T. Molecularpathways: targeting the cyclin D-CDK4/6 axis for cancer treatment. Clin.Cancer Res 21, 2905–2910 (2015).

50. Yamauchi, J., Miyamoto, Y., Sanbe, A. & Tanoue, A. JNK phosphorylation ofpaxillin, acting through the Rac1 and Cdc42 signaling cascade, mediates neuriteextension in N1E-115 cells. Exp. Cell Res. 312, 2954–2961 (2006).

51. Spiczka, K. S. & Yeaman, C. Ral-regulated interaction between Sec5 and paxillintargets exocyst to focal complexes during cell migration. J. Cell Sci. 121,2880–2891 (2008).

AcknowledgementsWe are grateful to J. Odajima and P. Sicinski for providing Ccnd1� /� and Ccnd1þ /þ

immortalized MEFs, M. Hendrix and E. Seftor for R3327-50A cells, and J. Yamauchi,P. Crespo, X. Bustelo and N. Agell for plasmids. We thank Sonia Rius, Isabel Sanchez andMarc Tarres for technical assistance, Maria Santacana for IHC images, J. Egea, C. Gallegoand M. Aldea for stimulating discussions and members of the CYC and PATONC labsfor helpful considerations. The proteomics analyses were performed by Carolina de laTorre and Sılvia Barcelo-Batllori from the IDIBELL Proteomics Unit that belongs toProteoRed, PRB2-ISCIII, supported by grant PT13/0001. This work was funded bySpanish Ministry of Education and Science (BFU2013-42895) (CSD2007-00015), grantsBFU2015–71308 and BFU2013–50245-EXP from the Spanish Ministry of Economy andCompetitivity, Catalan Government (SGR-559), and Fondo de Investigaciones Sanitarias(PI10/00604). N.P. Fuste was supported by a predoctoral fellowship of University ofLleida-Fundacio Alicia Cuello de Merigo, R. Fernandez-Hernandez (FPI) and T. Cemeli(FPU) were supported by predoctoral fellowships from Spanish Ministry of Educationand Science, and C. Mirantes was supported by a predoctoral fellowship of Fondo deInvestigaciones Sanitarias.




Author contributionsN.P.F., R.F-H., C.M., T.C., E.G., F.F., N.P. and X.D. performed the experiments. N.P.F.,R.F-H, X.D. and E.G. analyzed the data. M.R. and N.C. provided cell-culture support.N.P.F., M.R., J.T. and N.C. contributed materials and helped with procedures. X.D. andE.G. conceived and designed the project. E.G. supervised and coordinated all experi-mental work. F.F. and E.G. wrote the manuscript. All authors contributed discussionsand comments on the manuscript.

Additional informationSupplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Competing financial interests: The authors declare no competing financial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

How to cite this article: Fuste, N. P. et al. Cytoplasmic cyclin D1 regulates cell invasionand metastasis through the phosphorylation of paxillin. Nat. Commun. 7:11581 doi:10.1038/ncomms11581 (2016).

This work is licensed under a Creative Commons Attribution 4.0International License. The images or other third party material in this

article are included in the article’s Creative Commons license, unless indicated otherwisein the credit line; if the material is not included under the Creative Commons license,users will need to obtain permission from the license holder to reproduce the material.To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/





http://npg.nature.com/reprintsandpermissions/

http://npg.nature.com/reprintsandpermissions/

http://creativecommons.org/licenses/by/4.0/


Documents

Cytoplasmic cyclin D1 regulates cell invasion and